On 11/29/2013 01:45 PM, Michal Hocko wrote:
On Wed 27-11-13 19:46:01, Vladimir Davydov wrote:
We should start kmem accounting for a memory cgroup only after both its
kmem limit is set (KMEM_ACCOUNTED_ACTIVE) and related call sites are
patched (KMEM_ACCOUNTED_ACTIVATED).
This should be
On Wed 27-11-13 19:46:01, Vladimir Davydov wrote:
> We should start kmem accounting for a memory cgroup only after both its
> kmem limit is set (KMEM_ACCOUNTED_ACTIVE) and related call sites are
> patched (KMEM_ACCOUNTED_ACTIVATED).
This should be vice-versa, no? ACTIVE is set after
On Wed 27-11-13 19:46:01, Vladimir Davydov wrote:
We should start kmem accounting for a memory cgroup only after both its
kmem limit is set (KMEM_ACCOUNTED_ACTIVE) and related call sites are
patched (KMEM_ACCOUNTED_ACTIVATED).
This should be vice-versa, no? ACTIVE is set after
On 11/29/2013 01:45 PM, Michal Hocko wrote:
On Wed 27-11-13 19:46:01, Vladimir Davydov wrote:
We should start kmem accounting for a memory cgroup only after both its
kmem limit is set (KMEM_ACCOUNTED_ACTIVE) and related call sites are
patched (KMEM_ACCOUNTED_ACTIVATED).
This should be
We should start kmem accounting for a memory cgroup only after both its
kmem limit is set (KMEM_ACCOUNTED_ACTIVE) and related call sites are
patched (KMEM_ACCOUNTED_ACTIVATED). Currently memcg_can_account_kmem()
allows kmem accounting even if only one of the conditions is true.
Fix it.
We should start kmem accounting for a memory cgroup only after both its
kmem limit is set (KMEM_ACCOUNTED_ACTIVE) and related call sites are
patched (KMEM_ACCOUNTED_ACTIVATED). Currently memcg_can_account_kmem()
allows kmem accounting even if only one of the conditions is true.
Fix it.
6 matches
Mail list logo